Targeting MTAP increases PARP inhibitor susceptibility in triple-negative breast cancer through a feed-forward loop

Abstract

Triple-negative breast cancer (TNBC) represents the most malignant subtype of breast cancer. The clinical application of PARP inhibitors (PARPi) is limited by the low frequency of BRCA1/2 mutations in TNBC. Here, we identified that MTAP deletion sensitized genotoxic agents in our clinical cohort of metastatic TNBC. Further study demonstrated that MTAP deficiency or inhibition rendered TNBC susceptibility to chemotherapeutic agents, particularly PARPi. Mechanistically, targeting MTAP that synergized with PARPi by disrupting the METTL16-MAT2A axis involved in methionine metabolism and depleting in vivo s-adenosylmethionine (SAM) levels. Exhausted SAM in turn impaired PARPi-induced DNA damage repair through attenuation of MRE11 recruitment and end resection by diminishing MRE11 methylation. Notably, brain metastatic TNBC markedly benefited from a lower dose of PARPi and MTAP deficiency/inhibition synergy due to the inherently limited methionine environment in the brain. Collectively, our findings revealed a feed-forward loop between methionine metabolism and DNA repair through SAM, highlighting a therapeutic strategy of PARPi combined with MTAP deficiency/inhibition for TNBC.

Keywords: Amino acid metabolism; Breast cancer; DNA repair; Oncology; Therapeutics.

Figure 1. MTAP deficiency confers vulnerability to DNA damage–inducing agents.

(A) Flowchart illustrating the enrollment of patients with TNBC with distant metastasis, excluding brain metastasis, who were treated with DNA damage–inducing agents. NP, vinorelbine plus platinum; GN, gemcitabine plus vinorelbine. (B) Percentage change from baseline of 16 distant metastatic TNBC following DNA damage–inducing agents treatment. PD, disease progression; SD, stable disease; PR, partial response; CR, complete response. (C) Abdominal MRI showing the liver metastases (top) and pulmonary CT scan depicting the lung metastases (bottom) of the patient that achieved CR before and after treatment. (D) Schematic representation of TNBC PDX model establishment. (E) Copy number variant of MTAP in the 5 TNBC analyzed by whole-genome sequencing. (F and G) Representative immunohistochemistry staining (F) and Western blot (G) of MTAP in the 5 TNBC. Scale bar: 20 μm. (G) The experiment was repeated 3 times, and representative blots are presented.

Figure 2. MTAP deficiency/inhibition renders cells susceptible to PARPi.

(A) Predicted sensitivity to the indicated genotoxic agents for breast cancer with MTAP WT or deletion. (B–D) Tumor growth curves (B), tumor growth inhibition (C), and Kaplan-Meier survival curves (D) of PDX3 and PDX4 models treated with vehicle, olaparib (50 mg/kg), or veliparib (50 mg/kg). (E–H) Tumor growth inhibition and Kaplan-Meier survival curves of HCC70 (E and F) and PDX4 (G and H) xenograft models treated with olaparib (50 mg/kg), veliparib (50 mg/kg), or MTAPi (10 mg/kg), or the indicated combinations. (B, C, E, and G) Data are shown as the mean ± SD from 1 representative experiment of 5 mice per group. n = 10 mice per group in D, F, and H. P values are indicated. Significance was determined using (A) 2-sided Mann-Whitney U, (B) 2-way ANOVA, (C) unpaired t, (D, F, and H) log-rank (Mantel-Cox), or (E and G) 1-way ANOVA test.

Figure 3. PARPi regulates MAT2A intron retention by mediating METTL16 phosphorylation within Ser419.

(A and B) Representative Western blots (A) and quantitation (B) showing levels of MAT2A and METTL16 in HCC70 cells expressing control or METTL16 sgRNAs with or without cycloleucine (20 mM) treatment. (C) Schematic representation of RT-qPCR assay primers designed for total, intron retention (RI), or mature MAT2A mRNA. (D) RT-qPCR analysis of total, intron retention, and mature MAT2A mRNA levels in HCC70 cells with or without METTL16 KO treated with vehicle or cycloleucine (20 mM). (E–H) Representative Western blots and quantitation showing levels of MAT2A and METTL16 in HCC70 cells treated with the indicated dose of olaparib for 24 hours (E and F) or the indicated time period of olaparib at 5 μM (G and H). (I) RT-qPCR analysis of total, intron retention, and mature MAT2A mRNA levels in HCC70 cells treated with olaparib (5 μM), or cycloleucine (20 mM), or their combination. (J and K) Representative Western blots (J) and quantitation (K) showing levels of MAT2A and METTL16 in HCC70 cells expressing the indicated vectors treated with olaparib (5 μM), cycloleucine (20 mM), or their combination. (L) Schematic representation of PARPi downregulation of MAT2A expression by attenuation of its mRNA splicing through METTL16 conformational change induced by phosphorylation within Ser419. (A, E, G, and J) The experiment was repeated 3 times, and representative blots are presented. (B, D, F, H, I, and K) Data are shown as the mean ± SD from 3 independent experiments. P values are indicated. Significance was determined using (B, D, F, H, I, and K) 1-way ANOVA test.

Figure 4. PARPi and MTAPi in combination effectively kill tumor cells by reducing SAM.

(A) Schematic representation of the methionine recycling and salvage pathways. (B) Principal component analysis (PCA) score scatter plots of vehicle and olaparib combined with MTAPi groups. (C and D) Waterfall plot (C) and volcano plot (D) of intracellular metabolite level analysis using untargeted LC-MS in HCC70 cells treated with vehicle or olaparib (2 μM) combined with MTAPi (1 μM). (E and F) Intracellular SAM (E) and SAM/SAH ratio (F) of HCC70 cells treated with olaparib (2 μM), veliparib (2 μM), or MTAPi (1 μM) or the indicated combinations. (G–J) Intratumor SAM and SAM/SAH ratio of HCC70 (G and H) and PDX4 (I and J) xenograft tumors treated with olaparib (50 mg/kg), veliparib (50 mg/kg), or MTAPi (10 mg/kg) or the indicated combinations. Data are shown as the mean ± SD from 1 representative experiment of 5 mice per group. (K–M) Intracellular SAM (K), SAM/SAH ratio (L), and colony formation assay (M) of HCC70 cells treated with olaparib, veliparib, or MTAPi or the indicated combinations and supplemented with 100 μM Met, SAM, SAH, Hcy, or MTA. (E, F, and K–M) Data are shown as the mean ± SD from 3 independent experiments. P values are indicated. Significance was determined using (E–M) 1-way ANOVA test.

Figure 5. PARPi combined with MTAPi markedly inhibit tumor growth through diminishing SAM.

(A) Representative Western blots showing levels of MAT2A and MTAP in HCC70 cells treated with olaparib (2 μM), veliparib (2 μM), MTAPi (1 μM), or SAM (100 μM), or the indicated combinations. The experiment was repeated 3 times, and representative blots are presented. (B–I) Tumor growth inhibition, Kaplan-Meier survival curves, intratumor SAM, and SAM/SAH ratio of HCC70 (B–E) and PDX4 (F–I) xenograft tumors treated with olaparib, veliparib, or MTAPi or the indicated combinations. Olaparib and veliparib were used at 50 mg/kg and MTAPi was used at 10 mg/kg, intraperitoneally, 5 times per week. SAM was administered at 10 mg/kg subcutaneously each day. (B, D, E, F, H, and I) Data are shown as the mean ± SD from 1 representative experiment of 5 mice per group. n = 10 mice per group in C and G. P values are indicated. Significance was determined using (B, D, E, F, H, and I) 1-way ANOVA test or (C and G) log-rank (Mantel-Cox) test. (J) Schematic representation of PARPi exhibiting synergistic effect with MTAP inhibition.

Figure 6. SAM depletion caused by MTAP and PARP inhibition attenuates DNA repair.

(A–D) Representative micrographs and quantitation of neutral comet assay (A and B) and micronucleus assay (C and D) in the indicated cells without treatment or recovery at 0 hours, 6 hours, or 36 hours after olaparib (20 μM) treatment. (A and C) The experiment was repeated 3 times, and representative micrographs are presented. Scale bar: 50 μm (A); 10 μm (C). (E) Schematic representation of the DR-GFP reporter system. (F) Relative HR repair efficiency in the indicated cells. shBRCA1 was used as a positive control. (B, D, and F) Data are shown as the mean ± SD from 3 independent experiments. P values are indicated. Significance was determined using (B, D, and F) 1-way ANOVA test.

Figure 7. SAM depletion attenuates DNA repair by impairing methylation of MRE11-mediated end resection and recruitment.

(A–H) Representative micrographs and quantitation for RAD51 (A and B), RPA32 (C and D), BrdU (E and F), or MRE11 (G and H) foci formation in the indicated cells without treatment or recovery at 1 hour after olaparib (20 μM) treatment. (I) Representative Western blots immunoblotted with the indicated antibodies in whole-cell lysate and chromatin fraction of the indicated cells without treatment or recovery at 1 hour after olaparib (20 μM) treatment. (J) Representative Western blots showing methylation levels of MRE11 in cells with the indicated treatment. ADMA, asymmetric dimethyl-arginine antibody. (K) Schematic representation of a feed-forward loop between low SAM levels resulting from treatment with PARPi, MTAPi combined with MR, and PARPi-induced DNA damage. (A, C, E, G, I, and J) The experiment was repeated 3 times, and representative micrographs/blots are presented. (B, D, F, and H) Data are shown as the mean ± SD from 3 independent experiments. P values and scale bars are indicated. Significance was determined using (B, D, F, and H) 1-way ANOVA test.

Figure 8. Methionine restriction augments the combination efficacy of MTAP deficiency/inhibition and PARPi.

(A) Representative Western blots showing levels of MAT2A in HCC70 cells treated with olaparib (2 μM), veliparib (2 μM), or methionine restriction (MR) (20%) or the indicated combinations. The experiment was repeated 3 times, and representative blots are presented. (B–D) Intracellular SAM (B), SAM/SAH ratio (C), and colony formation assay (D) of HCC70 cells with or without MTAP deletion treated with olaparib, veliparib, or MR or the indicated combinations. (E–G) Intracellular SAM (E), SAM/SAH ratio (F), and colony formation assay (G) of HCC70 cells treated with olaparib, veliparib, MTAPi, or MR or the indicated combinations. (B–G) Data are shown as the mean ± SD from 3 independent experiments. P values are indicated. Significance was determined using (B–G) 1-way ANOVA test.

Figure 9. MR boosts the antitumor effect of MTAP and PARP inhibition in vivo.

(A–H) Tumor growth inhibition, Kaplan-Meier survival curves, intratumor SAM, and SAM/SAH ratio of HCC70 (A–D) and PDX4 (E–H) xenograft tumors treated with olaparib, MTAPi, MR, or the indicated combinations. All drugs were used at 10 mg/kg, intraperitoneally, 5 times per week. The control diet contained 0.86% methionine, and methionine restriction diet contained 0.12% methionine. (A, C–E, G, and H) Data are shown as the mean ± SD from 1 representative experiment of 5 mice per group. n = 10 mice per group in B and F. P values are indicated. Significance was determined using (A, C–E, G, and H) 1-way ANOVA or (B and F) log-rank (Mantel-Cox) test. (I) Schematic representation of MR enhancing the combination effect of PARPi and MTAP deficiency/inhibition through depletion of SAM.

Figure 10. MTAP deficiency confers a therapeutic vulnerability to lower dose of PARPi in brain metastatic TNBC.

(A) Schematic representation of treatment scheme for brain metastatic TNBC. (B–D) Representative in vivo bioluminescence imaging (B), quantification of radiance (C) prior to treatment or 21 days after treatment, and Kaplan-Meier survival curves (D) of intracardiac injection mouse models of MTAP WT or deleted HCC70 aggressive BrM cells, treated with vehicle or veliparib (10 mg/kg). (E) Schematic representation of intracranial injection of HCC70 luciferase or PDX cells. (F–H) Representative in vivo bioluminescence imaging (F), quantification of radiance (G) prior to treatment or 18 days after treatment, and Kaplan-Meier survival curves (H) of intracranial injection mouse models of MTAP WT or deleted HCC70 aggressive BrM cells, treated with vehicle or veliparib (10 mg/kg). (I) Kaplan-Meier survival curves of PDX3 and PDX4 intracranial injection mouse models treated with vehicle or veliparib (10 mg/kg). (C and G) Data are shown as the mean ± SD from 1 representative experiment. n = 12 mice per group in D, H, and I. P values are indicated. Significance was determined using (C and G) 1-way ANOVA or (D, H, and I) log-rank (Mantel-Cox) test.

Figure 11. MTAPi combined with low-dose PARPi profoundly benefits brain metastatic TNBC.

(A–C) Representative in vivo bioluminescence imaging (A), quantification of radiance (B) prior to treatment or 21 days after treatment, and Kaplan-Meier survival curves (C) of intracardiac injection mouse models of MTAP WT HCC70 aggressive BrM cells, treated with vehicle, veliparib (10 mg/kg), or MTAPi (10 mg/kg) or their combination. (D–F) Representative in vivo bioluminescence imaging (D), quantification of radiance (E) prior to treatment or 18 days after treatment, and Kaplan-Meier survival curves (F) of intracranial injection mouse models of MTAP WT HCC70 aggressive BrM cells, treated with vehicle, veliparib (10 mg/kg), or MTAPi (10 mg/kg) or their combination. (G) Kaplan-Meier survival curves of PDX4 intracranial injection mouse models treated with vehicle, veliparib (10 mg/kg), or MTAPi (10 mg/kg) or their combination. (B and E) Data are shown as the mean ± SD from 1 representative experiment. n = 12 mice per group in C, F, and G. P values are indicated. Significance was determined using (B and E) 1-way ANOVA or (C, F, and G) log-rank (Mantel-Cox) test.